August 2019
Intermediate to advanced
342 pages
9h 35m
English
As we have seen, the purpose of the algorithms is to learn to make correct predictions by generalizing from the training samples. All the algorithms, as a result of this learning process, manifest a generalization error that can be expressed as the following formula:

By Bias, we mean the systematic error made by the algorithm in carrying out its predictions, and by Variance, we mean the sensitivity of the algorithm to the variations affecting the analyzed data. Finally, Noise is an irreducible component that characterizes the data being analyzed.
The following diagram shows different estimators that are characterized ...
Read now
Unlock full access